Background

There are important international differences in cancer survival (Richards M. EUROCARE-4 studies bring new data on cancer survival. Lancet Oncology 2007, 8: 752-753) and the explanation for these is not known. The International Cancer Benchmarking Partnership (ICBP) was formed to explore our understanding of these differences. The ICBP is a unique and innovative international partnership of clinicians, academics and policymakers, the first of its kind, seeking to determine how cancer survival varies between jurisdictions and to explore reasons that could explain any differences. It has the following key objectives:

• Identifying specific causes of performance differences between countries with high quality cancer data
• Generating insights to improve cancer survival outcomes
• Laying the foundation for an ongoing benchmarking partnership to enable the countries to track the relative successes of their reform programmes and share good practice

The scope of the work covers primary cancers and focuses on four cancer sites: breast, lung, colorectal and ovarian cancer. Breast, colorectal and lung cancer were chosen as common cancers that contribute a large share of the burden of cancer disease in developed countries. Significant international differences in survival have been reported for these cancers.

This is a truly international partnership which involves 12 jurisdictions in 6 countries in 3 continents:

• Australia (New South Wales and Victoria)
• Canada (Alberta, British Columbia, Manitoba, Ontario)
• Denmark
• Norway
• Sweden
• United Kingdom (England, Northern Ireland, Wales)

The ICBP covers five areas of research (modules). Each module looks at different aspects of cancer survival to study possible reasons for differences between partners.

• Module 1: Epidemiology 
• Module 2: Population awareness and beliefs about cancer
• Module 3: Beliefs, behaviours and systems in primary care
• Module 4: Root cause of diagnosis and treatment delays 
• Module 5: Treatment, co-morbidities and other factors

Module 1: Epidemiology

Initial results from Module 1 were published in the Lancet in January 2011 (Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995—2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data) and were based cancer registry data on over 2.4 million cancer patients diagnosed during the period 1995 to 2007. Whilst survival in all countries improved for all four cancers studied and higher in Northern Ireland than elsewhere in the UK it was lower than that of the Australian and Canadian jurisdictions and two Swedish regions. Whilst the actual survival was still low, the rate of improvement was greatest in the UK. The differences in survival between jurisdictions suggested that later diagnosis or differences in treatment, particularly in Denmark and the UK, and in patients diagnosed aged 65 years and older were contributory factors. Further more detailed analysis has been carried out in each of the four cancer sites with particular emphasis on the role of stage at diagnosis in explaining differences in cancer survival. A preliminary paper was prepared on the comparability of stage data between countries ‘Comparability of stage data in cancer registries in six countries: lessons from the International Cancer Benchmarking Partnership’ in June 2012.

Key Findings

• Problems with the comparability of stage data internationally were highlighted in all four papers; the consistency and intensity of recording stage information varies widely between countries in the partnership across all cancer types studied.
• The UK was the worst at recording stage at diagnosis – across all four cancer types.
• The UK has the lowest overall cancer survival across four cancer types when compared to the other countries. Sweden was one of the best performing countries; people in Sweden were the most likely to survive at least one year after their diagnosis for breast, bowel and lung (Sweden did not submit data relating to ovarian cancer).
• In comparison to the other ICBP countries, Denmark generally had the fewest number of women diagnosed at earlier stages.
• Factors such as differences in treatment or access to treatment, differences in how doctors stage cancers and late diagnosis may partially explain the international cancer survival differences.
• Some of these factors are currently being investigated by other areas of the partnership.

Module 1 investigated at what role treatment plays in international survival differences. Results from this analysis are expected in late 2013.

Module 4 is looking at the time it takes between a patient noticing something was wrong to contacting a doctor, to then being diagnosed and starting treatment for the cancer. Researchers think that shorter times to diagnosis and treatment are likely a key element in improving survival for patients. First results of the study are expected in late 2014.

Ovarian cancer (Stage at diagnosis and ovarian cancer survival: Evidence from the International Cancer Benchmarking Partnership)

This study found that differences in stage at diagnosis may only partially explain the variation between countries in the proportion of women who survive ovarian cancer. The study found that the UK and Denmark had the lowest one-year ovarian cancer survival of the five countries. Denmark also had the fewest women diagnosed at earlier stages which could partly explain why their ovarian cancer survival is lower than in some other countries. Overall, the UK had similar proportions of women diagnosed at each stage of the disease as in the other participating countries. But women who were diagnosed at later stages of the disease were less likely to survive ovarian cancer in the UK. This suggests that treatment or access to treatment is poorer in the UK for women diagnosed at later stages, than in the other countries in the study.

This study also found that there was a large difference in records that were missing information on stage at diagnosis for ovarian cancer. The UK was the worst at recording stage at diagnosis. Three out of ten ovarian cancer tumours reported in the UK had no information on stage at diagnosis. This is compared to less than one out of ten ovarian cancers in Norway. It isn’t clear if this is because fewer women have the stage of their cancer recorded by doctors in the UK or whether this is because the transfer process of stage data to the cancer registries is less complete in the UK than in the other countries. Survival in the UK was also lower among those whose stage at diagnosis had not been recorded. The paper highlights the need for routine recording and transferral of stage data to cancer registries in all countries.

Lung cancer (Lung cancer survival and stage at diagnosis in Australia, Canada, Denmark, Norway, Sweden and the UK: a population-based study, 2004–2007)

Researchers found that differences in stage at diagnosis may only partially explain the differences between countries in the number of people who survive lung cancer. One-year small cell and non-small cell lung cancer net survival varied widely between these six countries in 2004-7. One-year survival from non-small cell lung cancer ranged from 30% in the UK to 46% in Sweden. Survival from this type of lung cancer was relatively low in Denmark, intermediate in Norway and higher in Australia and Canada. The UK also had the lowest one-year small cell lung cancer survival, Sweden and Australia had the highest.

The proportion of patients diagnosed at an early stage for non-small cell lung cancers was slightly lower in the UK and Denmark in comparison to other countries. Differences in the proportions of people diagnosed at different stages of the disease, ‘stage distribution’, could be due to delays in diagnosis or differences in staging. Denmark had low survival for patients with early stage disease, but average survival for those with more advanced disease. The UK survival figures were among the lowest at all stages compared to the other countries. In Canada patients had high survival at early stages of lung cancer but relatively poor survival at advanced stages of lung cancer.

There are many possible reasons for the international differences in survival including differences in treatment or access to treatment, late diagnosis or differences in how doctor's 'stage' lung cancers. Inaccurate or incomplete recording of stage information could lead to inappropriate treatment and poorer lung cancer survival. This could also affect the pattern of stage distribution in a country. This study found large differences in the number of records missing information on stage at diagnosis for lung cancer. Again the UK was the worst at recording stage. Three out of ten non-small cell lung cancers and four out of ten small cell lung cancers reported in the UK had no information on stage at diagnosis recorded. It isn’t clear if this is because fewer people are staged by doctors in the UK or because the transfer process of stage data to the cancer registries is less complete in the UK than in the other countries.

Breast cancer (Breast cancer survival and stage at diagnosis in Australia, Canada, Denmark, Norway, Sweden and the UK, 2000-2007: a population-based study. British Journal of Cancer advance online publication 28 February 2013)

The number of UK breast cancer patients diagnosed in the early stages of the disease was similar to the number in Canada, Norway and Sweden (countries with higher cancer survival than the UK). Denmark had the lowest proportion of patients diagnosed at the earliest stage, three out of ten women, compared to four out of ten women in Canada (a country with higher survival). This could partly explain why Denmark’s breast cancer survival is lower than in some other countries, as finding breast cancers at an early stage improves the chances of survival. More than 99% of women when diagnosed at the earliest stage of the disease (stages 1 and 2) survived their disease for at least 3 years in all the countries included in the study. This finding suggests that all of the countries in the study are good at treating early stage breast cancer. However survival rates for all stages combined for women diagnosed in the UK were lower when compared to other countries in the study.

UK survival figures for women diagnosed with late stage breast cancer were significantly lower than for other countries in the study. For women diagnosed at a late stage, five out of ten women in the UK survived their disease for at least one year after diagnosis compared to seven out of ten women in Sweden (the country with the highest survival). Three out of ten women in the UK survived their disease for at least three years after diagnosis compared to four out of ten women in Sweden. This suggests that treatment or access to treatment is poorer in the UK for women diagnosed at later stages, than in other countries. Treatment and late diagnosis may partially explain the differences between countries in the number of women who survive breast cancer. Survival differences could also be impacted by inaccurate or incomplete recording of stage information which may lead to the patient receiving inappropriate treatment. This study found that there were large differences in the number of records missing information on stage at diagnosis for breast cancer. The UK was the worst at recording stage. One out of four breast cancers reported in the UK had no information on stage at diagnosis recorded. It isn’t clear if this is because fewer women are have their stage recorded by doctors in the UK or because the transfer process of stage data to the cancer registries is less complete in the UK than in the other countries. 

Colorectal cancer

Wide differences in international stage distributions and survival exist between the countries in this study. Survival and stage distribution for colon (large bowel) and rectal (back passage) cancer were investigated separately in this study. Colon cancer is more commonly diagnosed early (stage A) in Canada, at intermediate stages (stage B and C) in Sweden and the UK, and at an advanced stage (stage D) in Denmark. The stage distribution for rectal cancer was similar in Canada, Norway and Sweden – with 45-52% of cancers diagnosed at stage A and B, around 25-30% at an intermediate stage and 21-23% diagnosed at the latest stage. Denmark had the highest proportion of patients diagnosed at the latest stage, D. Colon and rectal cancer patients in the UK were less likely than in other countries to be diagnosed at the latest stage (19% stage D in the UK compared to 31% in Denmark). The UK had the highest proportion of colon and rectal cancers diagnosed at stage C, an intermediate stage, compared to the other countries. The study found that the UK and Denmark had the lowest one-year colon and rectal cancer survival of the six countries. Survival was intermediate in Norway and Canada for colon cancer and highest in Sweden and Australia. Survival was intermediate in Norway for rectal cancer and highest in Sweden, Canada and Australia
Treatment and late diagnosis may partially explain the differences between countries in the proportion of people who survive colon and rectal cancer.

Survival differences could also be impacted by inaccurate or incomplete recording of stage information which may lead to the patient receiving inappropriate treatment. Late diagnosis appears to be contributing to poor survival in colon and rectal cancers in the Denmark.

There was a large difference in how many records were missing information on stage at diagnosis for colon and rectal cancer. The UK was the worst at recording stage at diagnosis; three out of ten colon and rectal tumours had stage at diagnosis data missing. The abstract of the paper published investigating stage at diagnosis and colorectal cancer survival in the ICBP countries can be found here.

Module 2: Population awareness and beliefs about cancer

‌

‌

‌

‌

‌

‌

Module 3: Beliefs, behaviours and systems in primary care

Module 3 looks specifically at the role of primary care in diagnosing cancer.  It explores the attitudes and beliefs of General Practitioners (GPs) and also examines differences in primary care systems involved in diagnosing cancer.  The aim of module 3 is to understand international differences in primary care that may impact on cancer survival. Together with the analyses from the other ICBP modules, results from Module 3 will inform healthcare policy in the UK in order to improve the number of people surviving from cancer.

Method and Fieldwork 

The Primary Care Physician survey

An international group, led by Dr Peter Rose in England, has developed this survey. It will provide an insight into the contribution of variation in primary care management of symptoms to cancer outcomes. It consists of two parts:

A set of general questions relating to GP practice administration, access to diagnostics, training and education and availability of advice from secondary care.  Some jurisdictions will also collect data on questions which are relevant to specific local issues, for example detail on clinical guidelines.

Two patient vignettes. The physician is asked to make choices about management in response to the presentation of the patient. The patient may or may not have cancer and this will be revealed at the end of the vignette. The vignettes aim to draw out any differences in the approach of primary care physicians to patients with these symptoms.

200 GPs per jurisdiction will be were surveyed in each jurisdiction. The approach to sampling varied in each jurisdiction due to differences in the availability of the contact details for primary care physicians.

The survey was rolled out to all 11 jurisdictions. Data collection was completed in 2014 and results were published in May 2015.  The analysis of the patient scenarios showed a statistically significant correlation (p<0.05 or better) between readiness to investigate or refer to secondary care at the first phase of each scenario and cancer survival rates for that jurisdiction. No consistent associations were found between readiness to investigate and selected PCP demographics, practice or health system variables with the authors suggesting that Some jurisdictions may consider lowering thresholds for PCPs to investigate for cancer—either directly, or by specialist referral, to improve outcomes. To find out more about this study please click here

The study took a more in-depth look at differences in the healthcare system that might influence on the GP’s management of potential cancer patients. Researchers looked specifically at factors of the healthcare system that might impact upon GP behaviour and potentially provide insight into variation in international cancer survival. This study compared the healthcare systems of Australia, Canada, Norway, Sweden, Norway, Denmark and the UK. Comparisons were made using an existing framework developed by researchers.

The study identified a high degree of similarity between jurisdictions on factors such as financing, regulations and the role of the GP as a central figure in the treatment and care of patients. Differences were found in relation to the degree of centralisation over decision making, free movement of patients between primary care providers, access to secondary care, and the existence of patient list systems. The researchers highlighted that is not possible to make a causal link between any of these factors and cancer survival. Further research into these factors was recommended, including the impact of co-payments on the GP-patient relationship.

If you are interested in reading, ‘How might healthcare systems influence speed of cancer diagnosis: a narrative review’ this paper is available free of charge from the Social Science and Medicine journal.

 

‌

‌

‌

‌

‌

‌

Module 4

A study investigating the root causes of diagnosis and treatment delays in four cancer sites (breast, lung, colorectal, and ovarian).

Cancer survival in NI, although improving, lags behind that of many countries for lung, breast, colon, rectum and ovarian cancer (Coleman et al 2011). The reasons for poorer survival may lie with delayed diagnosis. Late diagnosis of cancer can be caused by late patient presentation or delays in primary or hospital care. In NI, 64% of lung cancer patients presented with late stage disease and only 11% were suitable for curative surgery (Bannon & Gavin 2009).  In addition, 43% of colon cancers present as emergencies, while 57% of screen age women with breast cancer present outside of screening. Earlier diagnosis is known to improve survival.

The variation of outcomes in cancer survival across countries with similar levels of spending and resources remains unexplained. There is a wide range of views but a lack of clear evidence as to why cancer outcomes vary and therefore about how resources can be used most effectively. Survival differences can arise from many factors including patient characteristics, nature of diagnostic and screening services, cultural and societal factors and quality of treatments available.

This project will explore differences in total delay in the pathway to treatment for four cancers (breast, lung, colorectal and ovarian) and the contribution of different elements to that delay (e.g. patient, primary care, and hospital delays). Data collected will allow international comparison of the NI results to the 11 other jurisdictions.

Johnny Price	Consultant Gynaecologist	Belfast Trust
Stephen Kirk	Consultant Breast Surgeon	South Eastern Trust
Roy Maxwell	Consultant Colorectal Surgeon	Belfast Trust
Jonathan McAleese	Consultant Clinical Oncologist (lung)	Northern Trust
Miriam McCarthy Secretary: Amy Harvey	Consultant Public Health Medicine	Public Health Agency
Nigel Hart	GP	Dept. General Practice, QUB
Sarah Williamson/ Debbie Wightman	Cancer Services General Manager	Belfast Trust
Heather Monteverde	General Manager	Macmillan Cancer Support
Claire Leathem	NICRN Primary Care Nurse	Primary Care Network
Dalrene Masson	Clinical Audit Facilitator	GAIN
Eileen Dillon	Lead Clinical Research Nurse	NI Cancer Trials Centre
Ethna McFerran	Clinical Research Nurse	NI Cancer Trials Centre
Ruth Hall	Clinical Research Nurse	NI Cancer Trials Centre
Anna Gavin	Director	NICR
Conan Donnelly	Statistician	NICR
Jackie Boylan	Research Fellow: Project Coordinator	NICR

 

It aims to collect survey data from 800 patients and their General Practitioners to determine reasons for delay in cancer presentation and to document patient experiences during the process of diagnosis and treatment. Data is being collected using standardized methodologies and questionnaires which will allow comparison of the NI results to the 11 other jurisdictions (New South Wales, and Victoria (Australia), Alberta, British Columbia, Manitoba and Ontario (Canada), Sweden, Denmark, Norway, England, and Wales) as part of an international cancer benchmarking partnership.

 

‌

‌

‌

‌

‌

‌